Cree XP-L, XM-L performance in an XP size package

I did some tests of the XP-L V6 2C on a Noctigon in a P60 pill w/Nanjg at 3.85A (3 extra 7135's), using a LT2 host with an XML SMO reflector -- unimpressive. I wanted to compare head to head with a XM-L2 U2 1A, so, I first measured lumens only with the XM-L2 star, then swapped the star using the same pill/driver. Surprisingly, the initial results came out dead even @30 secs. I didn't use any wrapping on the P60 intentionally. I used healthy LG HE2's to ensure amp readings of 3.92A, and all my tests were done at that amp reading for both the XM-L2 U2 1A and the XP-L V6 2C.

I'm not happy with how equal the test configurations are because of:

  • XP-L in a XML SMO reflector
  • initial test done with the same low profile LED alignment piece as used for the XM-L2, so the reflector depth should match, but can't guarantee the LED centering, though it looks pretty good. The Noctigon shows around the LED because of the reduce size of course.
  • Maybe didn't wait long enough between tests - heat issues?
  • I left the thermal grease on the pill top as-is. I think I used a thicker grease that doesn't spread out, at the time - not sure if this had any negative effect, but should be equal for the XM-L2 and XP-L?

Throw was slightly better, sure within tolerances, but I got 20 kcd on the XP-L. I didn't do "before" measurements today, but previously measured 19 kcd.

Though the results are surprising low according to the specs, this could be a confirmation of Dale's tests, though I'm using the V6 bin. Still, the 2C tint is pretty nice for getting the same output of a XM-L2 U2 1A.

I did later try this for the XP-L: http://intl-outdoor.com/cree-xpe-xpg-insulation-gaskets-10-pcs-p-487.html, in white, and got slightly better #'s in throw and lumens -- very slight.

Here's the test setup/results:

  • Using L2T host - low profile SS bezel, AR lens from CNQ, copper braided tail spring
  • Using HE LG2 cells at full charge
  • P60#1 pill - Nanjg @3.85A, XM-L2 U2 1D/SinkPAD, braided + spring (no wrap)
  • P60#2 pill - Nanjg @3.85A, XM-L2 U2 1A/Noctigon, braided + spring (no wrap)
  • L2T/P60 #1: 3.89A tail, lumens: 1,207 @start, 1,125 @30 secs
  • L2T/P60 #2: 3.92A tail, lumens: 1,258 @start, 1,156 @30 secs
  • L2T/P60 #2 swapping in XP-L VC 2C/Noctigon: 3.92A tail, lumens: 1,251 @start, 1,156 @30 secs, throw: 20 kcd
  • same as above but with disc LED piece: 3.92A tail, lumens: 1,258 @start, 1,163 @30 secs, throw: 21 kcd

Thanks Tom.

So as suspected it’s basically just XM-L2 cut down. Still, it should be interesting for triples.

..and interesting for small lights running at less than 3A (as long as the new higher binned xml2 are not available).

+1

With the disappointing results when overdriven, we're going to have to get our Russian friend to develop XM-L2 triple and quad boards to fit whichever Carclo makes for the XP-L. Carclo just fit over the die, not the substrate, right?

Put an XP-L V6 2C on a 10mm SinkPAD in a T10 (AA light), thread here: https://budgetlightforum.com/t/-/27321. I really like it. Easier to do than an XM-L2 because I didn't have to mod the reflector, and it really cranks lumens in the nice 2C tint, holds the output better, and I think it doesn't get as hot quickly -- no measurements to back that up though... Not sure if an XM-L2 would be any less, of course...

The dome seems to be cut out or recessed above the die, like a convex lens. As such, it seems to be taped or glued on around the edges. It’s not stuck to the die face itself. So in less than 2 hours the entire intact dome will flip right off and leave a clean die face like none I’ve done before. Easy, quick, very clean. And the 2A has a nice warm tint after de-doming, perhaps a slight green to it but consistently (as much as 2 emitters can show) better than the 1A XM-L2 in the tint dept.

It will be interesting to see if/when other tint’s come out if there will be one that de-domes a neutral white, no green or blue or yellow or orange or whatever. I want that one!

Carclo is concerned that a quad 20mm optic for the XP-L will have too much flood and spill due to the large size of the dome unless they go bigger. Personally, I definitely want 20mm, and I want as much flood and spill as possible. How about you guys?

20mm optics for XP-L. YES please.

Even if it's a pure flooder?

ESPECIALLY if it’s a pure flooder! :slight_smile:

They use the same TIR for everything that will fit in the hole. (So does Ledil) I guess this would be a big step away from that since it would basically be a big-hole TIR just for the XP-L / XM-L / etc.

I have no idea what the market for one would be if it’s pure flood.

I can understand their concern. I know I want it, and so do at least a couple others here so far. If they're still hesitant, I may suggest a kickstarter for it, then at least those of us interested can pre order it and they'll take less risk.

Their main focus is fixed-lighting, I think it's just a coincidence that some of their parts happen to work acceptably well in our flashlights. I don't think they're gonna ever be interested in making parts specifically for flashlights.

They've offered to do a custom order before. I never followed up on the cost.

Maybe the cost of doing small-run (1000 units?) would pleasantly surprise us.

That's a last resort option, and one I hope that they'll do a kickstarter for instead of making us do a group buy. I figure if they do a kick starter, they're more likely to let their other clients know instead of having us flashlight enthusiasts go it alone.

You betcha! They’ll sell tons!

The important question is, are you okay with that 20mm XP-L optic having lots of flood and spill? It would be even more floody than you'd expect when going from XP-G2 to XM-L2.

Depending on use, flood can be the imperative. I can see it being very useful as a headlight, not so much for fine work, but larger areas, like in a mine/cave or something. Hunting through an unlit shed/room for a tool/item.

But I gather mules can do so too, Im not sure what the benefit would be over nothing except protection of the components and appearance and a mule could have plain glass lens. Not that appearance is nothing, in fact to me its somewhat critical.